A differential refractometer - Analytical Chemistry (ACS Publications)

R. O. Butterfield and H. J. Dutton. Analytical Chemistry ... Historical Glimpses of Analytical and Quality Assessment Methods for Fats and Oils. Herbe...
2 downloads 0 Views 348KB Size
A Differential Refractometer D. RAU

AND

W. E. ROSEYEARE, University of Wisconsin, Madison, Wis.

T

bands. The outside cell, shown in the lower part of Figure 3, is made by cutting a half cylinder from a bottle 8 om. in diameter and 9.5 cm. long and cementing a piece of plate glass on each end. The only precautions necessary in making the cells is that the ends of the outside cell be parallel within 0.1" and the line of intersection of the two optical faces of the triangular cell be perpendicular within 0.3" to the surface of the channel iron. The plate-glass top which supports the triangular cell is held against the projecting end of the rectangular cell by a spring at the other end, preventing rotation as well as movement toward or away from the light source. The small prism, P, containing water is made by cementing two pieces of thin plate glass (1.5 X 3 cm.) together with a separator a t one end. The angle between the faces is 2' 55'. The axis of this prism must be parallel to that of the cell prism within 0.3". A shutter placed at 0 can be turned up or down to cut out the li ht from either the upper or lower parts of the triangular cell. %oth light paths are left open except when the solution is so very dilute that the two sets of bands overlap. With pure water in both compartments, the readings through the upper and lower compartments should be identical. If these readings are different, it is due to a strain in the triangular cell.

HIS refractometer was designed to be a rapid and reliable instrument for determining concentrations of solutions too dilute to permit the use of the Pulfrich or immersion refractometers. It makes use of the bending of a ray of light when passed through a rectangular cell containing a triangular compartment, the former being filled with pure solvent and the latter with solution. This part of the instrument is similar in principle t o the Ketteler (2) and HaberLowe (I) refractometers, but is much more practical because of the new method of observing and measuring the deflection and of determining the zero reading. The complete instrument is shown in Figure 1. The upper diagram in Figure 2 shows the essential parts of the refractometer, while the lower diagram shows the path of the light ray. The achromatic lens, L, projects through the cell an enlarged image of the slit, Sp, at point F , 110 om. from the lens. This image is broken up into three bright interference bands by the double slits, 88, which are 0.31 cm. (0.125 inch) wide and 0.31 cm. (0.125 inch) apart. The appearance of the interference bands and cross hairs is shown in the circle. These interference bands are observed by the microscope, M , having cross hairs in the eye piece. Readings are taken by means of the position of a small-angle water prism, P, in a glass cell which slides on a 60-cm. steel metric scale. This prism is moved along the scale until the interference bands are brought back to the zero position. It does not displace the interference bands at all when a t the focal point of the microscope, but when moved toward the cell it dis laces the image near1 in proportion to the distance move$ from the zero point, $. The mounting for this prism has guides that slide along the sides of the steel scale and prevent it from rotating more than 0.02'. Larger rotations would affect the readings. It is made as thin as possible. The collimator of a spectroscope serves very well for the lens, L, and slit, SZ,if the edges of the slit are ground smooth on an oil stone. The lens has a focal length of 20 om. The light source, J , is a 100watt opal-glass electric light bulb. The larger slit, 81, protects the end of the collimator tube from the heat of the lamp. The collimator tube supporting Sz is surrounded by a larger tube to protect it from drafts which would make the readings erratic. A Gaertner MlOl microscope is suitable to observe the interference bands. All parts of the refractometer are mounted on a piece of IO-cm. (4-inch) channel iron 125 cm. long.

The first step in adjusting the instrument is t o fill the rectangular cell with distilled water, leaving the triangular cell out for preliminary adjustments. Open slit Xp wide and move it toward or away from the lens until an image of it is projected at point F on a piece of white paper. Close the slit slowly, observing the image of the slit through the microscope. When the slit can just barely be seen, refocus the microscope on this image. Interference bands should then appear. The quality of the bands can be improved by slight chan es in the width of the slit and position of the microscope. Ifhen sharp interference bands are obtained, replace the triangular cell and fill it with water. Look a t the slit from point 0. If the light coming from both double slits does not appear the same, move the triangular cell until it lets both beams of light ass through it equally well and make a final readjustment o? the microscope to give the best interference bands. Table I shows readings taken every minute for 14 minutes, after filling the triangular cell with a solution of 0.05 N sodium chloride. The last reading was taken after leaving the solution in the cell until the next day. The readings became constant in 5 minutes after filling the cell. The reading at 10 minutes is an accidental error of observation due to eye fatigue, from looking continuously at the interference bands. Eliminating this value and the first four taken while the temperature was being equalized, the average deviation from the mean is 0.042 mm. or an average error of 0.032 per cent for this 0.05 N solution of sodium chloride. For a 0.1 N solution the average error would be one-half as great, or 0,016 per cent. However, the error does not decrease below this value at higher concentrations because the bands become colored through dispersion of the salt.

FIGURE1. COMPLETE INSTRUMENT The triangular cell, shown in the up er part of Figure 3, is built on the under side o?a piece of late glass, having a 2-cm. round hole in the mixdle. The optical faces are pieces of ordinary commercial plate glass and extend 1 cm. below the bottom of the triangular cell, which is 1.2 cm. deep and has an inside length of 8.5 om. The angle between the faces is 137". One piece of heavy platinum foil is used for the bottom and back of the cell because of its higher heat conductivity. A narrow slot 1.5 mm. deep is cut in the plate glass so that the platinum foil is mortised into the glass. This prevents the cement from covering much of the optical surface and gives it much greater strength. This cell requires 5 cc. of liquid in order to transmit bright interference

0

s.s.

*/

P

AND PATHOF LIGHTRAY FIGURE2. REFRACTOMETER

72

JANUARY 15, 1936

ANALYTICAL EDITION

The zero reading changes considerably, owing to warping of the instrument by temperature changes and relieving of strains, but the difference between the zero reading and the solution reading remains constant. Therefore it is always necessary to take a zero reading immediately before the solution reading. It is important to keep the instrument out of drafts when in use or the zero reading will change so rapidly that the reading will be erratic. If the cell is opened for only a few seconds, sufficient evaporation occurs to cool the liquid so that one must wait several minutes after covering again for the temperature to equalize. Table I1 shows the effect of changing the position of the triangular cell in the rectangular compartment. The only small displacement that would affect the reading is moving the triangular cell toward the light source. Since the glass plate is pressed against one end of the rectangular cell, it can always be replaced so that the reading will be affected less than 0.03 mm., a negligible error. TABLEI. READINGS Time Min. 1 2 3 4

Zero Reading

Solution Reading

Difference

0.08 0.04 0.04 0.06 0.07 0.07 0.09 0.12 0.10 0.10 0.11 0.12 0.13 0.14 0.25

13.30 13.28 13.28 13.29 13.28 13.28 13.30 13.33 13.32 13.36 13.32 13.33 13.36 13.35 13.47

13.22 13.24 13.24 13.23 13.21 13.21 13.21 13.21 13.22 13.260 13.21 13.21 13.22 13.21 13.22

Cm.

6

6 7 8

9 10 11 12

a

13 14 Next day Accidental error.

POSITIONOF CELL TABLE11. EFFECTOF CHANGING Prism Position Normal position 0 . 6 O t o right 0.5O to left, 2 mm. to side 6 mm. toward light Original position

ZeSo Reading 1.77 2.27 0.86 1.67 1.76 1.61

Solution Reading 14.98 15.48 14.07 14.88 15.16 14.72

Difference 13.21 13.21 13.21 13.21 13.40 13.21

FIGURE 3. DIAGRAM OF CELLS convenient for analytical purposes to calibrate the instrument directly in terms of concentration of the solution to be analyzed. If the solution has a refractive index below that of the reference liquid, the prism, P , is reversed on the scale. If absolute differences of refractive indices are desired, it is easier to calibrate the instrument with two solutions of known refractivity than to measure the angles of the prisms. This refractometer has been used for transference number determinations. For the small concentration changes involved, the concentration was proportional to the scale reading, within 0.1 per cent. Using this 137” cell, An = 0.002003 for a 60-cm. or maximum scale reading. This would be the reading given by a 0.23 N sodium chloride solution. If more concentrated solutions are to be analyzed, they may be diluted or a prism with a smaller angle may be used.

Summary The sensitivity of the instrument increases rapidly as the angle of the triangular cell is increased, but angles larger than 137” are not recommended since the bands become weaker and more diffuse as the angle is increased. The water prism, P , gives the instrument more than twice the range that it, would have with a prism of ordinary glass. The higher dispersion of the water compensates considerably for the dispersion due to the solution in the refractometer. A scale reading of 60 cm. corresponds to a difference in refractive index of about 0.002 or a deviation of about 0.5” by the cell. For such small angles, the displacement of the movable prism is practically proportional to the difference in refractive index. The scale readings for certain differences in refractive indices were calculated, using eight-place trigonometricai functions, for the mean of the two beams which give the interference bands. The difference in refractive index, with water as the reference liquid, is given within one part in ten thousand by the formula A% =

(33355 s

- 2.83 8 2 ) x

10-9

where X is the scale reading in centimeters. The second term amounts to only 0.1 per cent for a 13-cm. scale reading and for smaller readings it may be neglected. For the maximum scale reading of 60 cm., the second term amounts to 0.5 per cent. Since the difference in refractive index in general is not exactly proportional to the concentration, it is more

A differential refractometer has been developed that uses white light and has a sensitivity of 5 X lo-’. It has been used to determine the concentrations of aqueous solutions with refractive indices from 0.0004 to 0.002 greater than pure water with an accuracy of 0.1 per cent. This corresponds approximately to 0.04 to 0.2 N sodium chloride solutions. The scale readings are nearly a linear function of the difference in refractive indices. Readings can be taken much more easily and rapidly than with the liquid interference refractometer. Literature Cited (1) Haber and Ltiwe, 2. Elektrochem., 13, 460 (1907); 16, 37 (1910). (2) “Handbuch der Physik,” Vol. XVIII, p. 675, Berlin, Julius Springer, 1927. RECEIVED May 29, 1935.

In the paper on “A Modified Persulfate-hseCORRECTION. nite Method for Manganese” by E. B. Sandell, I. M. Kolthoff, and J. J. Lingane [IND. ENG.CHEM.,Anal. Ed., 7, 256 (1935)J the 15th line from the bottom in the first column of page 258 should read “Dissolve 1.2 to 1.3 grams of pure arsenic trioxide-,” instead of “Dissolve 2.5 grams of pure arsenic trioxide-.” I. M. KOLTHOFF